Technical Artifacts for "JATA: Separating Retrieval Coverage, Evidence Fitness, and Evidence Admission in Multi-Step Retrieval-Augmented Generation"
Abstract
Technical artifact package accompanying the IEEE Access manuscript “JATA: Separating Retrieval Coverage, Evidence Fitness, and Evidence Admission in Multi-Step Retrieval-Augmented Generation” (Manuscript ID: Access-2026-34157). The archive contains frozen experimental results, reproducibility code, scorer-forensics materials, corpus and cluster manifests, author-adjudicated evidence-fitness annotations, matched-admission evaluation artifacts, external QASPER evidence-selection results, figure source data, provenance records, package manifests, and a claim-to-artifact map. The package supports the analyses reported in the manuscript, including the 1,884-record correctness-scorer forensic audit, the 630-pair duplicate-session stability analysis, the 30-question evidence-fitness adjudication, the matched-admission experiment, and the external QASPER evidence-level validation. Restricted third-party source text is not redistributed. Where redistribution restrictions apply, the archive provides stable identifiers, provenance information, hashes, cluster assignments, and reconstruction information instead. The archive is designed to be immutable under the documented reproduction procedure. The reproduction script regenerates the reported headline results without modifying indexed package contents.
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.